TalkChess.com

AMD's next-generation Ryzen chips, due to hit the shelves in a few months, should have excellent performance-per-watt, possibly significantly better than Intel's offerings. Watch the first 4 or five minutes of this video: https://www.youtube.com/watch?v=g39dpcdzTvk These chips will almost certainly ...

What am I missing? I don't see any way to force the black king off the c-file w/o dropping the d3 pawn (which appears to actually loose!)

Larry, anyone.
Halving the cpu resources should cost the typical engine somewhere around 20(?) elo? Yet MPV =2 kills ~100 points. Why is the MPV performance hit so high? (A simple question with - I suspect - a complex answer! )

Here are some benchmarks. As suggested, be sure to get a card capable of the running the fp-16 back end, as seen in the config notes. "--backend=cudnn-fp16"

https://docs.google.com/spreadsheets/d/ ... 1508569046

I'm not 100% sure this is happening, but it seems that setting MultiPV to N causes the engine to search all N first moves to a more or less equal depth. This short-circuits the engine's finely tuned pruning parameters and theoretically weakens the engine slightly. But the engine is also significantl...

petero2 wrote: ↑

Fri Feb 08, 2019 5:44 pm

I have uploaded stockfish_cluster_win64.7z to this public dropbox folder:

Thank you for the compile and specific instructions!

Thanks Mr. X

What card were you running before? What NPS did it generate?
Have you enabled the FP16 back end on the new card? What NPS are you getting now?

GPU memory bandwidth doesn't really matter. It's mostly cuda cores * clock speed. And you can use multiple GPUs as long as they are roughly the same. like a 2080 and 2080 ti or a 1060 and 1070. Up to 4 GPUs works well now. Now that the 2060 is out there's really no reason not to get that for Leela ...

Speculation over the R3K series is even more wild and wooly than usual for a new CPU lineup. This fellow has some interesting information, leaks, analysis, etc. https://www.youtube.com/watch?v=ReYUJXHqESk